The development and assessment of spectrum management methods for the copper access network are usually conducted under the assumption of accurate channel information. Acquiring such information implies, in practice, estimation of the crosstalk coupling functions between the twisted-pair lines in the access network. This type of estimation is not supported or required by current digital subscriber line (DSL) standards. In this work, we investigate the impact of the inaccuracies in crosstalk estimation on the performance of dynamic spectrum management (DSM) algorithms. A recently proposed crosstalk channel estimator is considered and a statistical sensitivity analysis is conducted to investigate the effects of the crosstalk estimation error on the bitloading and on the achievable data rate for a transmission line. The DSM performance is then evaluated based on the achievable data rates obtained through experiments with DSL setups and computer simulations. Since these experiments assume network scenarios consisting of real twisted-pair cables, both crosstalk channel estimates and measurements (for a reference comparison) are considered. The results indicate that the error introduced by the adopted estimation procedure does not compromise the performance of the DSM techniques, that is, the considered crosstalk channel estimator provides enough means for a practical implementation of DSM.